AI-POWERED LIGHTWEIGHT FRAMEWORK FOR ANOMALY DETECTION AND DDOS PREVENTION IN SDN
Abstract
The current dated of your focus on networking systems has been extraordinarily supportive to frequent field such as education, medicine, finance, government, etc. It has also been observed that there is an growing demand for dependable, swift, and productive automated systems. As a result, there is a increasing interest in, and broad application of, SDN. Like other networking systems, SDNs allow for central control of networked devices, setting them apart from traditional networking systems SDNs hold the advantage of programmability and custom control. However, freedoms of programmability still retain fewer security challenges and make SDN systems more appealing to certain methods of cyberattacks, particularly the distributed denial of service. So This Paper presents a lightweight framework for detecting and mitigating Distributed Denial of Service (DDoS) attacks in Software-Defined Networking (SDN) environments. SDN enhances network management but introduces vulnerabilities that make it susceptible to DDoS attacks. The framework includes flow collection, feature extension, anomaly detection, and mitigation modules. The Naïve Bayes model achieved 93.67% accuracy, with a recall of 1.00 and precision of 0.91. The logistic regression model showed 97.08% accuracy, with a recall of 0.99 and precision of 0.97. The framework was validated using Mininet and the Ryu controller, with traffic data collected via the SDN controller. This framework contributes to network security by offering an effective solution for DDoS detection and mitigation in SDN environments. Future work will enhance the mitigation module and refine the user interface.
Keywords: NLP, AI, ML, DDOS, Mitigating Framework, Network Security
